This small grant requests funds for a pilot study designed to determine if within individual variability in degree of cyclical dysphoric mood in women is related to within individual variability in patterns of steroid hormones. The major focus of this investigation will be the within individual relationship between intensity of cyclical dysphoric mood and the absolute levels and rates of change of the ovarian steroids. In addition, endocrine data from a normal control group will be available for comparison of steroid levels and rates of change between women with and without strong cyclical mood change associated with the menstrual cycle. The aims of this project will be accomplished by monitoring changes in dysphoric mood and ovarian steroids during the course of three consecutive menstrual cycles in women with prospectively documented changes in mood associated with the menstrual cycle. The subjects will maintain a structured diary and collect a sample of the first morning urine void daily. The diary will be used to determine each individual's most and least symptomatic menstrual cycles which will be defined in terms of the degree of change in dysphoric mood from an asymptomatic period in the mid follicular phase to the late luteal phase. Urine samples will be assayed for the major end products of estradiol, progesterone, and luteinizing hormone. The steroid metabolites measured will include the major quantitative metabolites of the three primary metabolic pathways of estradiol metabolism, and a major metabolite of progesterone. Data analyses will compare the most and least symptomatic menstrual cycles from subjects whose degree of mood change during their most symptomatic menstrual cycle was at least 15 percent greater than the degree of mood change during their least symptomatic cycle. The within subject endocrine comparisons will include levels of estrogen and progesterone production during the luteal phases, rates of increase of estrogen and progesterone during the early and mid luteal phases, and rates of decrease of estrogen and progesterone during the late luteal phases. The results of these analyses will provide information about the degree to which variation in ovarian steroid patterns is associated with variation in the intensity of dysphoric moods in a sample of women with significant mood cyclicity. A second set of analyses will compare women with perimenstrual syndrome and normal controls in terms of reproductive hormone levels and rates of change.